CCHHEEMMIICCAALL EENNGGIINNEEEERRIINNGG TTRRAANNSSAACCTTIIOONNSS

VOL. 39, 2014

A publication of

The Italian Association

of Chemical Engineering

www.aidic.it/cet Guest Editors: Petar Sabev Varbanov, Ji Jaromr Kleme, Peng Yen Liew, Jun Yow Yong

Copyright 2014, AIDIC Servizi S.r.l.,

ISBN 978-88-95608-30-3; ISSN 2283-9216 DOI:10.3303/CET1439079

Please cite this article as: Tohneanu M.C., Pleu V., Iancu P., Bumbac G., Bonet Ruiz A.E., Bonet Ruiz J., 2014,

Simulation and process integration of clean acetone plant, Chemical Engineering Transactions, 39, 469-474

DOI:10.3303/CET1439079

469

Simulation and Process Integration of Clean Acetone Plant

Mdlina C.Tohneanu*a, Valentin Pleua, Petrica Iancua, Gheorghe Bumbaca,

Alexandra Elena Bonet Ruizb, Jordi Bonet Ruizb aUniversity POLITEHNICA of Bucharest, Centre for Technology Transfer in Process Industries (CTTPI), 1, Gh. Polizu

Street, Bldg A, Room A056, RO-011061 Bucharest, Romnia bUniversity of Barcelona, Department of Chemical Engineering, 1, Mart i Franqus Street, 6th Floor, Room 664, E-

08028 Barcelona, Spain

pres2014@chim.upb.ro

Acetone is produced worldwide by cumene hydroperoxide process. Alternatively, free of aromatic

compounds acetone is obtained by isopropyl alcohol (IPA) dehydrogenation. The feedstock is azeotropic

mixture IPA-water (Luyben, 2011). This contribution aims to evaluate process feasibility both in terms of

reaction engineering and separation by distillation. Rational use of energy is based on process integration

A process flowsheet with better performance is proposed. The dehydrogenation reaction is endothermic

with number of moles increase. Higher temperature and decreased IPA partial pressure using inert (water)

are favourable. Reaction kinetics in vapour phase is simulated in Aspen HYSYS with Langmuir-

Hinshelwood-Hougen-Watson (LHHW) model, considering both direct reaction and reverse reaction terms,

based on experimental data (Lokras, 1970). Such more realistic model is not considered in other papers

(Luyben, 2011). The chemical reactor is simulated as an ideal CSTR. High volatile and non-condensable

compounds from the cooled reactor effluent are separated into a flash unit. To minimize the loss of

acetone in gas stream, an absorption column with water is considered. Possible separation schemes for

liquid effluent (consisting of acetone, IPA and water) are evaluated with residual curves maps (RCM) built

in SIMULIS

ver.1.3. This system has IPA-water azeotrope, a saddle node. The boundary, separes two

distillation regions. Both direct and reverse separation schemes are possible. Two distillation columns are

used to separate the components of flash unit bottom stream product and absorbtion column bottom

stream product. As designed by RCM, for direct scheme, acetone separates in first column top product,

and IPA-water azeotropic mixture separates as second column top product. This is recycled to meet plant

feedstock stream. Water separated as second column bottom stream product is partly recycled, after

cooling, to the absorption column. To minimise utilities consumption and to synthesize process heat

exchangers network (HEN), maximum energy recovery (MER) methodology is used with SPRINT

software. The proposed HEN is included in process flowsheet. Energy saving is 35 % for hot utilities and

37 % for cold utilities, compared to non-integrated scheme.

1. Introduction

Clean acetone is used in pharmaceutical formulations and as solvent for biologic sensitive processes. A

new process flowsheet based on IPA dehydrogenation is taken into account. The purpose of this paper is

to simulate acetone from IPA process flowsheet. Systematic and realistic modeling, as well as energy

savings by process integration, is considered with appropriate computer tools. The idea of the process

flowsheet is based on block diagram given by Turton et.al (1998). Substantial improvement is considered

both in reaction engineering aspects, and in separation feasibility. Systematic approach based on onion

diagram is applied to develop the process flowsheet. Realistic reaction kinetic model, based on

experimental data is considered. Separation scheme is obtained with RCM diagram, built in SIMULIS

ver.1.3. Process flowsheet data extracted is introduced in SPRINT

for process integration. Grand

Composite Curve is used to select utilities. Energy saving solution for HEN synthesis is based on MER

methodology.

470

2. Process Description

The azeotropic fresh feed (0.65 mole fraction IPA-0.35 mole fraction water) and the recycled unreacted

mixture IPA/water are mixed feedstock. To reach reactor temperature of 598.15 K a preheat train is used,

as the reaction is endothermic. The reactor effluent, containing acetone, hydrogen, water and unreacted

IPA is cooled severely and then a flash unit separates phases. The non-condensable gas stream is

absorbed with water to recover entrained acetone. With reasonable effort, about 67 % is recovered from

gas stream. Absorber bottom stream product and flash unit liquid stream are mixed before the separation

section. As indicated by RCM diagram, for direct scheme, the first column separates acetone as top

product and the second column separates IPA-water azeotropic mixture as top product. The azeotrope

obtained from the second column is recycled. Second column bottom product (water) is partially recycled

to absorption column. Keeping in mind physical-chemical processes, more fluid packages are used. For

vapor phase processes, as reaction, Peng-Robinson Twu fluid package is used. For liquid phase

processes, the nonideality is described with NRTL fluid package. Binary coefficients are given by Aspen

HYSYS data bank.

Figure 1. Process flow diagram for acetone production

3. Reaction kinetics

An onion diagram approach involves the reactor as starting point for flowsheet development (Kleme et al,

2010). The temperature values for the inlet (598.15 K) and outlet (623.15 K) streams are proposed in

literature for catalyst optimal operation (Turton et al, 1998). The dehydrogenation reaction is endothermic,

with a standard heat of reaction of 56 MJ/kmol. For this reason the reactor heating is needed (heating oil is

used). The pressure, assumed to be constant, is 235 kPa. A CSTR reactor model is used to predict the

performance of a fluidized-bed reactor for IPA dehydrogenation. Single-pass conversion of 90 % is

typically indicated in literature (Turton et al, 1998), under above mentioned conditions. Reactor simulation

provides 80 % conversion for the process flowsheet developed in this paper. The reaction is kinetically

controlled and occurs over catalyst surface. Reaction kinetics is modeled with (LHHW) model, containing

both direct reaction term and reverse reaction term. In Table 1 main catalyst characteristics (SiO2

471

supported Cu) from experimental data (Lokras, 1970) are shown. The reaction rate expression and kinetic

model parameters are given below (Lokras, 1970):

HHIPAIPA

p

ACHIPA

CKCK1

K

CCCk

r

, kmol/(m3h) (3)

T

6.3606exp4907.230k , reaction rate constant, kmol/(kgh) (4)

T

7.2685exp101688.0K 2IPA , adsorption equilibrium constant for alcohol, m

3/kmol (5)

T

7.12666exp109536.9K 12H , adsorption equilibrium constant for hydrogen, m

3/kmol (6)

Table 1. Catalyst characteristics

Surface area 31.01 m2/kg

Apparent density 2481.80 kg/m3

Bulk density 850.10 kg/m3

Reactor feedstock and effluent composition are given in Table 2.

Table 2. Reactor feed and effluent stream composition

Composition Reactor Feed Stream Effluent Stream

xIPA 0.6500 0.1153

xWater 0.3500 0.2365

xAC 0.0000 0.3240

xH 0.0000 0.3240

4. Feasibility Thermodynamics of Separation by Distillation

A systematic approach aimed to propose innovative topology for separation scheme is used, based on

advanced tools (Garcia Jurado et al, 2013). Detailed analysis of physical behaviour is accomplished.

Effluent stream (consisting of acetone, IPA and water) separation in high purity fractions is suppressed by

IPA-water azeotropic mixture. Using RCM built in SIMULIS possible separation schemes are evaluated.

The system presents a saddle node for IPA-water azeotropic mixture. From RCM shape it is obvious that

both direct and reverse separation schemes are feasible. Presenting distillation columns position, in Figure

2 and Figure 3 those possibilities are illustrated. However, the direct scheme looks more favorable,

because the top product of the first column is a quite pure (96 %) acetone, and the top product of the

second column is the azeotropic mixture IPA-water. Consequently, in this paper, the direct scheme is

considered for process flowsheet systematic development. The second column, as shown in Figure 2,

separates two streams: the azeotropic mixture as top product and water as bottom product.

5. Separation section

5.1 Flash Unit (V-101) Reactor effluent stream components are separated based on flash vaporisation as volatilities are very

different. Flash unit is assumed to be isothermal. The inlet and outlet temperatures are 288.15 K. The

pressure assumed constant is 185 kPa. Flash separation is essentially for this case, because the effluent

stream is split into two streams: hydrogen enriched gas stream, and acetone enriched liquid stream.

5.2 Absorber unit (T-101)

Inlet streams temperature is 288.15 K. Column pressure is considered constant. To recover entrained

acetone from non-condensable gas stream 142.5 kmol/h water is needed. Fresh water flowrate is 7.4

kmol/h, the rest is provided by recycled water from bottoms of second distillation column. Outlet streams

composition of flash unit and absorber are given in Table 3.

472

Figure 2. Direct separation scheme

Figure 3. Reverse separation scheme

Table 3. Flash unit and absorber outlet streams composition

Composition Vapor Stream

(Flash)

Liquid Stream

(Flash)

Off-Gas

(Absorber)

Bottom Product

(Absorber)

xIPA 0.0042 0.1753 0.0001 0.0034

xWater 0.0068 0.3604 0.0169 0.9632

xAC 0.0641 0.4641 0.0220 0.0334

xH 0.9249 0.0002 0.9610 0.0000

5.3 Distillation columns (T-102 and T-103) Absorber bottom stream product and flash unit liquid stream are mixed. As resulted from Figure 2, two

distillation columns are used to separate the resulted ternary mixture. As the first column has partial

473

condenser, three specifications are needed: acetone recovery degree in distillate, reflux ratio and

hydrogen molar flow in gaseous distillate stream. The column produces 96 % acetone purity, for 50,000 t/y

acetone productivity. The bottom product contains IPA (0.1476 mole fraction) and water. This stream is

then separated into IPA-water azeotrope and water streams in the second column. Two specifications are

needed for the second column : IPA recovery degree in distillate, and distillate composition of water. The

second column waste product is partially recycled as absorption column feed.

6. Recycle and Adjust

Recycling contributes to raw materials consumption decrease. Waste products or unreacted feed recycling

has become an important element of process sustainability. Two recycle operations are considered: IPA-

water azeotropic mixture recycle and water recycle. Water recycle stream is adjusted as target variable

meet required specifications. In this case, water recycle stream is set at 135.1 kmol/h and the target

variable, the absorption column inlet water, is set at 7.4 kmol/h. Another Adjust operation is required on

fresh feed stream. Raw material mass flowrate is adjusted to 136.4 kmol/h. Therefore, entire azeotrope

amount is recycled.

7. Process Integration

Pinch analysis is used to determine potential for energy saving. Heat exchanger network synthesis is

based on MER methodology. The first step is data extraction from non-integrated process flowsheet. Data

is introduced in SPRINT. The minimum difference temperature value is set to 10 K, based on heuristics.

Figure 4. Grand Composite Curve

Utilities selection is based on Grand Composite Curve. Figure 4 illustrates suitable utilities: heating oil and

low pressure steam are used as hot utilities, boiling feed water, cooling water and shaft water as cold

utilities. New HEN is presented in Figure 5, temperatures are in C, built for better economic feasibility.

The proposed HEN is included in process flowsheet. Table 4 presents utilities consumption in four different

situations.

Table 4. Utilities consumption analysis

Utility Minimum

consumption New HEN

Simulation of

Integrated Scheme

Simulation of Non-

Integrated Scheme

Hot 8.01 MW 8.23 MW 10.26 MW 13.70 MW

Cold 5.45 MW 5.67 MW 7.14 MW 11.38 MW

474

Figure 5. New heat exchangers network

8. Conclusion

This paper presents the results of isopropanol dehydrogenation process simulation. Study of designing a

two-column distillation configuration to separate acetone, IPA and water is accomplished. The proposed

scheme (direct scheme) provides good purity acetone. Unreacted feedstock is recycled. Bottom water

from second distillation column is recycled to absorption column. Acceptable return amount of recycled

water is determined with Adjust operation. Energy saving solution is found, based on Pinch analysis.

Energy saving is 35 % for hot utilities and 37 % for cold utilities, compared to non-integrated scheme.

Nomenclature

C molar concentration, kmol/m3

D - distillate product molar flow, kmol/h

F - feed molar flow, kmol/h

Kp - chemical equilibrium constant, kPa

W - waste product molar flow, kmol/h

Subscripts

AC - Acetone

H - Hydrogen

References

Garcia Jurado M.B., Pleu V., Bonet Ruiz J., Bonet Ruiz A.E., Tuluc A., Llorens Llacuna J., 2013,

Simulation of a hybrid reactive extraction unit. Biodiesel synthesis, Chemical Engineering Transactions,

35, 205-210, DOI:10.3303/CET1335034.

Kleme J., Friedler F., Bulatov I., Varbanov P., 2010, Sustainability in the Process Industry: Integration and

Optimization, McGraw-Hill Inc., USA.

Lokras S.S., Deshpande P.K., Kuloor N.R., 1970, Catalytic Dehydrogenation of 2-Propanol to Acetone,

Ind. End. Chem. Process Des. Develop., 293-297.

Luyben L.W., 2011, Design and Control of the Acetone Process via Dehydrogenation of 2-Propanol, Ind.

Eng. Chem. Res., 50, 12061218.

Turton R., Bailie R.C., Whiting W.B., Shaeiwitz J.A., 1998, Analysis, Synthesis and Design of Chemical

Processes, 1st ed., Prentice Hall: Upper Saddle River, New Jersey, USA, 789-796.